Machine Learning Engineer

You will be a key driver in our transition from reactive repairs to a proactive service model.

By leveraging high-frequency instrument telemetry and service data, you will build the predictive analytics directly impacting how we support our global customer base.

Key Responsibilities

Proactive Solution Development: Design and deploy ML models to predict component failures and estimate Remaining Useful Life (RUL).

Anomaly Detection: Build robust algorithms to detect silent deviations in machine performance from high-frequency sensor and log data.

Data Product Integration: Collaborate with the team to consume and refine data products from Snowflake and Databricks.

End-to-End ML Pipelines: Develop, test, and scale ML pipelines on Databricks/Snowflake.

Scalable MLOps: Own the end-to-end lifecycle of the models—from experimentation in notebooks to production deployment and monitoring using MLflow.

Actionable Insights: Work with domain experts to ensure model outputs are not just "scores," but clear, actionable steps for field engineers.

GenAI Collaboration: Support the integration of predictive insights into our GenAI-solutions, helping provide context-aware troubleshooting steps based on model outputs.

Technical Skills

The Essentials: Mastery of Python and SQL. Proficiency in PyTorch/TensorFlow.

ML Foundations: Proven experience with classical ML (XGBoost, Random Forest, Scikit-learn) applied to Time-Series or Sensor data.

Predictive Modeling: Strong understanding of Time-Series Analysis, Survival Analysis, and Anomaly Detection (e.g., Isolation Forests, Autoencoders)

Big Data Ecosystem: Hands-on experience with Databricks/Spark for processing large-scale machine datasets.

Statistical Depth: Familiarity with Survival Analysis or reliability engineering concepts is a strong plus.

Nice to Have

Experience with Deep Learning architectures (e.g., LSTMs or GRUs) for sequential data.

Familiarity with GenAI/LLM integration (building tools or agents).

Knowledge of dbt for data modeling within the ML pipeline.